Determination of feature boundaries in a digital representation of an anatomical structure

Abstract

A virtual anatomical structure can be analyzed to determine enclosing three-dimensional boundaries of features therein. Various techniques can be used to determine tissue types in the virtual anatomical structure. For example, tissue types can be determined via an iso-boundary between lumen and air in the virtual anatomical structure and a fuzzy clustering approach. Based on the tissue type determination, a deformable model approach can be used to determine an enclosing three-dimensional boundary of a feature in the virtual anatomical structure. The enclosing three-dimensional boundary can be used to determine characteristics of the feature and classify it as of interest or not of interest.

Description

TECHNICAL FIELD The field relates to software analysis of images. BACKGROUND Technology for non-invasive observation of soft tissues of the body has provided significant advances in the field of medicine. For example, a number of techniques now make it possible to routinely image anatomical structures such as the heart, colon, bronchus, and esophagus within the body. The widespread availability of skilled technicians and reduction in cost of the necessary equipment has encouraged the use of non-invasive imaging as a part of routine preventive care. Non-invasive imaging reduces the risk of observation-related injury or complication and reduces discomfort and inconvenience for the observed patient. These advantages encourage patients to undergo more frequent screening and permits earlier detection of potentially life-threatening conditions. For example, malignant or premalignant conditions can be identified and diagnosed at an early stage, when treatment is more likely to be succes...

AI Technical Summary

Problems solved by technology

However, as the number of slices increases, the amount of data presented to the physician becomes more difficult to efficiently analyze.

Although progress has been made in employing software to assist in detection of anatomical features, there are significant limitations to the current automated techniques.

For example, one problem consistently plaguing such systems is the overabundance of false positives when detecting features of interest.

However, the software also tends to incorrectly identify too many structures as features of interest (i.e., the software exhibits poor specificity).

False positives are troublesome because any identified positives must be considered and evaluated by a human classifier (such as the physician or a technician).

Even if a feature can be quickly dismissed as a false positive, too many false positives consume an inordinate amount of time and limit the usefulness of the software-based approach.

Images